Presentation is loading. Please wait.

Presentation is loading. Please wait.

Presented by Xu Miao April 20, 2005

Similar presentations


Presentation on theme: "Presented by Xu Miao April 20, 2005"— Presentation transcript:

1 Presented by Xu Miao April 20, 2005
SIFT Presented by Xu Miao April 20, 2005

2 Outline Motivation of SIFT Scale-space extrema detection
Scale-space function Local extrema detection Detection sampling Keypoint localization Orientation assignment Keypoint descriptor Comparison of Harris-Laplacian and SIFT Image matching 5/18/2019 CSE577 Computer Vision II

3 Motivation of SIFT (copy from 576 slides)
Image content is transformed into local feature coordinates that are invariant to translation, rotation, scale, and other imaging parameters SIFT Features 5/18/2019 CSE577 Computer Vision II

4 Motivation of SIFT --Advantages of local features (copy)
Locality: features are local, so robust to occlusion and clutter (no prior segmentation) Distinctiveness: individual features can be matched to a large database of objects Quantity: many features can be generated for even small objects Efficiency: close to real-time performance Extensibility: can easily be extended to wide range of differing feature types, with each adding robustness 5/18/2019 CSE577 Computer Vision II

5 More motivation… (copy)
Feature points are used also for: Image alignment (homography, fundamental matrix) 3D reconstruction Motion tracking Object recognition Indexing and database retrieval Robot navigation … other 5/18/2019 CSE577 Computer Vision II

6 Scale-space extrema detection
Scale-space function The only reasonable one: Laplacian of Gaussian kernel is a good choice of scale invariance Difference of Gaussian kernel is a close approximate to scale-normalized Laplacian of Gaussian 5/18/2019 CSE577 Computer Vision II

7 Scale-space extrema detection
Gaussian is an ad hoc solution of heat diffusion equation Hence k is not necessarily very small in practice K has almost no impact on the stability of extrema detection even if it is significant as sqrt of 2 5/18/2019 CSE577 Computer Vision II

8 Scale Invariant Detection (Copy)
Common approach: Take a local maximum of this function (convolution of kernel and image) Observation: region size, for which the maximum is achieved, should be invariant to image scale. Important: this scale invariant region size is found in each image independently! f region size Image 1 f region size Image 2 scale = 1/2 s1 s2 5/18/2019 CSE577 Computer Vision II

9 Scale-space extrema detection
σ2^(1/s) σ S + 3 S+2 5/18/2019 CSE577 Computer Vision II

10 Scale-space extrema detection
Sampling in scale for efficiency How many scales should be used per octave? S=? More scales evaluated, more keypoints found S < 3, stable keypoints increased too S > 3, stable keypoints decreased S = 3, maximum stable keypoints found We could not evaluate the 5/18/2019 CSE577 Computer Vision II

11 Scale-space extrema detection
Pre-smooth before extrema detection is equivalent to spatial sampling Sigma is higher, # of stable keypoints is larger Lower sampling frequency preferred? To utilize the information smoothed off, first double the original image, which also increases the stable kepoints found by 4 5/18/2019 CSE577 Computer Vision II

12 Keypoint localization
Detailed keypoint determination Sub-pixel and sub-scale location scale determination Ratio of principal curvature to reject edges and flats (detect corners?) 5/18/2019 CSE577 Computer Vision II

13 Keypoint localization
Sub-pixel and sub-scale location scale determination 5/18/2019 CSE577 Computer Vision II

14 Keypoint localization
Reject flats: < 0.03 Reject edges: r < 10 Is it Harris corner detector? 5/18/2019 CSE577 Computer Vision II

15 Keypoint localization
832 233x189 536 729 5/18/2019 CSE577 Computer Vision II

16 Orientation assignment
Create histogram of local gradient directions at selected scale Assign canonical orientation at peak of smoothed histogram Each key specifies stable 2D coordinates (x, y, scale,orientation) 5/18/2019 CSE577 Computer Vision II

17 Keypoint descriptor Invariant to other changes (Complex Cell)
In experiment, 4x4 arrays of 8 bin histogram is used, in total of 128 features for one keypoint 5/18/2019 CSE577 Computer Vision II

18 Scale Invariant Detectors (copy)
Harris-Laplacian1 Find local maximum of: Harris corner detector in space (image coordinates) Laplacian in scale scale x y  Harris   Laplacian  SIFT (Lowe)2 Find local maximum of: Difference of Gaussians in space and scale scale x y  DoG  1 K.Mikolajczyk, C.Schmid. “Indexing Based on Scale Invariant Interest Points”. ICCV D.Lowe. “Distinctive Image Features from Scale-Invariant Keypoints”. Accepted to IJCV 2004 5/18/2019 CSE577 Computer Vision II

19 Scale Invariant Detectors(copy)
Experimental evaluation of detectors w.r.t. scale change Repeatability rate: # correspondences # possible correspondences K.Mikolajczyk, C.Schmid. “Indexing Based on Scale Invariant Interest Points”. ICCV 2001 5/18/2019 CSE577 Computer Vision II

20 Comparison with Harrison-Laplacian
Affine-invariant comparison Translation-invariant – local features Rotation-invariant Harrison-Laplacian PCA SIFT Orientation Shear-invariant Eigen values No Within 50 degree of viewpoint, SIFT is better than HL, after 70 degree, HL is better. 5/18/2019 CSE577 Computer Vision II

21 Comparison with Harrison-Laplacian
Computational time: SIFT is few floating point calculation HL uses iterative calculation which costs much more 5/18/2019 CSE577 Computer Vision II

22 Object recognition by SIFT keypoint matching
Efficient nearest neighbor algorithm Best-Bin-First (modification of k-d tree) Hough transformation to cluster features into 3-feature groups Solving affine parameters by pseudo-inversion to verify the matching model Final decision is made by Bayesian approach 5/18/2019 CSE577 Computer Vision II

23 5/18/2019 CSE577 Computer Vision II

24 5/18/2019 CSE577 Computer Vision II


Download ppt "Presented by Xu Miao April 20, 2005"

Similar presentations


Ads by Google